Benoit Blanchet

Objective Interindividual variability in response to rituximab remains unexplored in antineutrophil cytoplasmic antibody (ANCA)–associated vasculitides. Rituximab pharmacokinetics (PK) and pharmacodynamics (PD) as well as genetic polymorphisms could contribute to variability. This ancillary study of the MAINRITSAN 2 trial aimed to explore the relationship between rituximab plasma concentration, genetic polymorphisms in PK/PD candidate genes, and clinical outcomes. Methods Patients included in the MAINRITSAN2 trial ( ClinicalTrials.gov identifier: NCT01731561) were randomized to receive a 500‐mg fixed‐schedule rituximab infusion or an individually tailored regimen. Rituximab plasma concentrations at month 3 (CM3) were assessed. DNA samples (n = 53) were genotyped for single‐nucleotide polymorphisms within 88 putative PK/PD candidate genes. The relationship between PK/PD outcomes and genetic variants was investigated using logistic linear regression in additive and recessive genetic models. Results One hundred and thirty‐five patients were included. The frequency of underexposed patients (<4 μg/ml) in the fixed‐schedule group was statistically lower compared to that in the tailored‐infusion group (2.0% versus 18.0%; P = 0.02, respectively). Low rituximab plasma concentration at 3 months (CM3 <4 μg/ml) was an independent risk factor for major relapse (odds ratio 6.56 [95% confidence interval (95% CI) 1.26–34.09]; P = 0.025) at month 28 (M28). A sensitivity survival analysis also identified CM3 <4 μg/ml as an independent risk factor for major relapse (hazard ratio [HR] 4.81 [95% CI 1.56–14.82]; P = 0.006) and relapse (HR 2.70 [95% CI 1.02–7.15]; P = 0.046). STAT4 rs2278940 and PRKCA rs8076312 were significantly associated with CM3 but not with major relapse onset at M28. Conclusion These results suggest that drug monitoring could be useful to individualize the schedule of rituximab administration within the maintenance phase.